evolution spamassassin plugin defaults to local mode, spamassassin defaults to network mode

Bug #48872 reported by Chris Moore
4
Affects Status Importance Assigned to Milestone
evolution (Ubuntu)
Invalid
Medium
Ubuntu Desktop Bugs
spamassassin (Ubuntu)
Invalid
Low
Ubuntu Desktop Bugs

Bug Description

I installed spamassassin to use it with evolution. When I boot the system, /etc/init.d/spamassassin starts the 'spamd' daemon, with these flags:

    DOPTIONS="-d --pidfile=$PIDFILE"

That makes a spamd which listens on a socket for incoming connections.

However, the spamassassin plugin in evolution is set up to not be able to use such a spamd process by default. In evolution-2.6.1/mail/evolution-mail.schemas.in.in, we see the following:

    <schema>
      <key>/schemas/apps/evolution/mail/junk/sa/local_only</key>
      <default>true</default>
         <short>Use only local spam tests.</short>

That default 'true' value means that evolution's spamassassin plugin will only use a spamd process if it was run with the "-l" or "--local" flags. See evolution-2.6.1/plugins/sa-junk-plugin/em-junk-filter.c - there's code which checks for the --local flag:

    if (em_junk_sa_local_only && try_system_spamd) {
      argv [i++] = "ps ax|grep -v grep|grep -E 'spamd.*(\\-L|\\-\\-local)'|grep -E -v '\\ \\-p\\ |\\ \\-\\-port\\ '";

The upshot of this is that evolution ends up starting its own spamd process, with the --local flag, and I end up with two spamd processes running.

We should decide on a suitable default for both spamassassin and evolution - should spamd run with --local or not? It makes no sense for two spamd daemons to be running.

Revision history for this message
Ante Karamatić (ivoks) wrote :

This can/should be fixed in evolution. I don't see it as a spamassassin bug.

Changed in spamassassin:
status: Unconfirmed → Rejected
Revision history for this message
Chris Moore (dooglus) wrote :

The point is that the default setting is different in the two programs. This can/should be fixed in either one of them.

Your argument could be used unchanged by the evolution package maintainer.

Revision history for this message
Robert Collins (lifeless) wrote :

one possibility is to have a separate package that has the evolution spamassassin plugin *and* a configuration change for spamd to set it to --local mode.

A different alternative is to have the actual spamd available in a does-not-run-by-default package, with 'spamassassin-daemon' or something to have it run by default.

At the heart of it though, we need to get the two maintainers talking to each other. Its not technically hard to do!

Revision history for this message
Chris Moore (dooglus) wrote : Re: [Bug 48872] Re: evolution spamassassin plugin defaults to local mode, spamassassin defaults to network mode

The choice to be made is should spamassassin be fast (local checks
only) or should it be effective (using network checks).

Revision history for this message
Peter Whittaker (pwwnow) wrote :

Is this not a duplicate of https://launchpad.net/distros/ubuntu/+source/evolution/+bug/9870 ?

I'm reluctant to mark 48872 as a dupe, as others may feel they are separate problems, but from reading 9870, it seems to me that spam filtering has been mangled in Evolution for years, this isn't something new.

For example, https://launchpad.net/distros/ubuntu/+source/evolution/+bug/62192 also concerns broken spam handling.

To me, these seem to be symptons of the same underlying problem: Evolution's spam filter doesn't work out of the box, cannot be trained without counter-intuitive actions (mark as spam, then ham), and does not cooperate with user addins (which shouldn't be necessary, IMHO).

On that last note, https://launchpad.net/distros/ubuntu/+source/evolution/+bug/37878 seems to come from this same underlying problem.

Would the problem get more or less attention and TLC if these bugs were all marked dupe of a 2+ year old high importance confirmed bug?

Revision history for this message
Rich Johnson (nixternal) wrote :

Is this still and issue with any of the latest stable releases?

Changed in evolution:
status: Unconfirmed → Needs Info
Revision history for this message
Daniel Holbach (dholbach) wrote :

As described in the previous comments, your report lacks the information we need to investigate the problem further. We'll close this report for now - please reopen it if you can give us the missing information.

Changed in evolution:
assignee: nobody → desktop-bugs
importance: Undecided → Medium
status: Needs Info → Rejected
Revision history for this message
Chris Moore (dooglus) wrote :

What information would you like me to add?

As far as I can tell my report gives more than enough details to reproduce the problem.

I didn't answer the "is this still an issue" question because I no longer use ubuntu and so don't know the answer to that one, but I don't see that as a valid reason for rejecting the bug.

Revision history for this message
Sebastien Bacher (seb128) wrote :

We rather need somebody getting the problem to work on it, there is thousand of desktop bugs open and only few people working on them, there is no way to work on everything at the moment

Revision history for this message
Chris Moore (dooglus) wrote :

"we're too busy" isn't a reason to reject a valid bug report.

If my bug report was somehow incomplete, please tell me what more you need to know.

Revision history for this message
Sebastien Bacher (seb128) wrote :

Daniel asked if it still happens on a current Ubuntu, you didn't reply, that's why it has been closed. Reopening if you think that's still a problem, marking low importance, somebody needs to look at that then

Changed in spamassassin:
assignee: nobody → desktop-bugs
importance: Undecided → Low
status: Rejected → Unconfirmed
Changed in evolution:
status: Rejected → Unconfirmed
Changed in spamassassin:
status: Unconfirmed → Rejected
Revision history for this message
Pedro Villavicencio (pedro) wrote :

We are closing this bug report as it lacks the information, described in the previous comments from Daniel and Sebastien, we need to investigate the problem further. However, please reopen it if you can give us the missing information and don't hesitate to submit bug reports in the future.

Changed in evolution:
status: New → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.